Research Journal of Applied Sciences, Engineering and Technology 5(4): 1160-1168, 2013
ISSN: 2040-7459; e-ISSN: 2040-7467
© Maxwell Scientific Organization, 2013
Submitted: June 19, 2012
Accepted: July 04, 2012
Published: February 01, 2013
Implementation of Scheduling Algorithm with Robotic Arm and Analytical Plate for
Clinical Chemistry Analyzer
1
Sudha Ramasamy and 2Padma Thiagarajan
Department of Electrical Engineering, VIT University, Vellore-632014, India
2
Department of Bio Sciences and Technology, VIT University, Vellore-613402, India
1
Abstract: Complete automation is the ultimate goal in health care industry and this is of utmost importance in
clinical laboratories. Processor based bio robots are involved in all these automation procedures. In this study, the
indigenous robotic platform, used in clinical chemistry analyzers, which is highly flexible and user friendly for small
or medium sized hospitals, is designed. A microcontroller based robotic arm is put forth as the robotic platform and
this is capable of handling blood samples, reagents, etc. The basic design is of a compact, three circular analytical
plates, placed one inside the other. The samples and reagents are loaded onto the analytical plate; the arm then
transfers the blood samples and reagents successively to the reaction cell and if required, next to an incubating unit.
Two different arms perform all the different tasks based on the controlling algorithm. The circuit manipulating the
path of the robot arm, along with other controlling circuitry of the arm is embedded within the arm itself. By
automating this unit, the flexibility and throughput of the tests will be increased. Controlled and precise use of
reagents and high accuracy of results are additional advantages. Data handling is also simplified. The robotic arm
and analytical plate has been designed, a prototype model has been made and synchronization between the two has
been achieved. The clear description of arm and analytical plate movement along with the synchronization
algorithms are presented in this study.
Keywords: Analytical plate, automation, clinical chemistry analyzer, robotic arm, sampling system
INTRODUCTION
One of the different parameters of automation is
the high cost associated with the fabrication of
analytical instruments. In this study, the fabrication of a
low cost automated servomotor controlled robotic arm
is put forth to handle the samples and reagents from the
analytical plate to the testing unit. By increasing the
level of automation, accurate use of reagents and
samples is feasible leading to valid results. The use of
manpower is also minimized in clinical laboratories.
The automatic blood sampling unit is incorporated and
it is used only for measuring the blood glucose level.
RIT (Reduced Idle Time) algorithm and it carries the
job without job delay time. Even this requires the
further future improvements in interfacing, optimization
of the hardware components, and schedulers. Transfer
of samples and reagents in the micro plates and other
communicating devices are communicated to each other
and all the patient information can be recognized by the
Radio Frequency Identification (RFID) system. The
feasibility of the system is validated through its
preliminary experiments (Choi et al., 2011a, b). The
robotic arm controlled by a microcontroller and driven
by the stepper motor has 4 degrees of freedom. Choi
et al. (2006) have suggested the robotic platform for the
clinical tests in the hospitals which satisfied 70
analytical methods for clinical test and also reduced the
amount of reagents consumed. Nakamachi (2010) has
been presented with Self Monitoring Blood Glucose
device (SMBG) which comprises of automatic blood
sampling unit with blood glucose measurement.
Infrared imaging is employed in this methodology. The
major drawback of this method is that it is a prototype
best suited only for measuring the blood glucose and
not for carrying out other analytical clinical tests. Choi
et al. (2008) proposed the technique for retrieving the
patient information can be easily recognized through
the RFID and by making all the components easily
replaceable, the operation process and maintenance of
the system is improved. The pneumatic pipette tip
which is presented exclusively in each sample tube is to
avoid the cross contamination between the samples.
The proposed system design is capable of handling the
samples in fully automated storage and sample
retrieval, including entering and dispatching of stored
sample (Saitoh and Yoshimori, 2008). A compatible
system which is developed is capable of handling any
reagent kit which is available in the market can be used
for the testing purpose and this system is working based
on the principle of absorbance transmittance
photometry (Taneja et al., 2005). Bhatia et al. (1998)
Corresponding Author: Sudha Ramasamy, Department of Electrical Engineering, VIT University, Vellore - 632014, India
1160
Res. J. Appl. Sci. Eng. Technol., 5(4): 1160-1168, 2013
have been designed an expert system which can
perform the different iterative modifications to the
model until all the requirements are been satisfied. Er
(2001) proposed a Hybrid Adaptive Fuzzy Controller
(HAFC) for selectively compliance assembly robot arm
for the applications which are best suited with
industries. The major advantage of this technique is that
it does not require mathematical model. HPFC
prototype was successfully simulated and implemented
in real time which resulted with reduced development
time as well as the reduction in cost.
MATERIALS AND METHODS
The analyzer system comprises of four units, viz.:
•
•
•
•
Analytical Plate
Sampling System
Controlling unit
Measuring unit
The Analytical plate contains three circular plates, with
the innermost one containing the samples and the
middle plate holding the reagents. The outermost one
comprises of the reaction cells. Each of the 3 rotatable
plates is driven independently by servo motors with all
of them being controlled by a micro controller. The
sampling system has two arms. They are perfectly
synchronized with the analytical plate so as to pick up
the samples and reagents from inner and middle plate.
Rotation of analytical plate and moving of the arm is
programmed
and
controlled
through
the
microcontroller. All the reagent containers and sample
tubes have their own respective pipette tips. The
pipettor in the arm positions the pipette tip in first
reagent container and transfers it to the required number
of reaction cells. This is based on the number of
samples loaded in the sampling system. For N number
of samples and M number of reagents, the arm needs to
take up reagents totally M×N times. Each time when
the reagent and samples are loaded in the reaction cell,
the temperature will be maintained at 37°C. This is
achieved by incorporating the Peltier system around the
reaction cells. The stirring arm stirs each reaction cell
contents and transfers the content to micro flow cell and
goes for wash cycle. Once readings are recorded, the
micro flow cell is washed, next reaction mixture loaded
and similarly processed. In this study unique analytical
plate is designed and it consists of three circular plates
kept one inside the other. The rotation of each plate is
individually controlled by three independent servo
motors and programmed through microcontroller. The
design of sample and reagent containers tray are having
an exclusive pneumatic pipette tips mainly to avoid the
cross contamination between the samples and to
eliminate the wash cycle process. The number of
reagent containers corresponds to the number of tests,
which could be performed for each patient. Three
different robotic arms are incorporated. These arms are
intended to handle the sample, reagent, stirring the
reaction cells and transferring the contents to the micro
flow cell. The results will be read out from the micro
flow cell and stored in the memory. Arm and analytical
plate rotations are controlled by the ATMega2561
processor.
Analytical plate: Here an analytical plate containing 3
circular plates is proposed. The inner plate holds
containers for 24 samples along with six standards and
2 controls each with a capacity of 0.8 mL. The middle
plate is a reagent tray with 36 reagent containers each
with a capacity of 25 mL. This reagent tray is designed
to easily detachable and can be refrigerated. The outer
most circular plate has 128 reaction cells each with a
capacity of 1.2 mL. This reaction cell array is
maintained at 37°C. With these 128 reaction cells, at
least 4 different tests for each patient can be performed.
After performing four different tests for each patient the
reaction cell array goes into the wash cycle mode and
all the reaction cells are cleaned. Then the arm starts the
process again for the fifth test. Each sample tube
contains the blood samples and a pipette tip inside. The
tip can be easily locked with the pneumatic pipette tip
which is attached to the arm. The same arrangement is
made for the reagent containers also. The reagent arm
picks up one particular reagent and puts it in the
reaction cell. The number of times this action needs to
be performed can be decided by the number of samples
loaded in the sampler.
Likewise the sampler arm needs to take the
samples from each sampler tube and put it in the
reaction cells. By accessing each sample tube by the
sampler arm, all the sample tubes are loaded by the
pipette. Figure 1 shows the analytical plate with three
different circular rotations. To complete 36 tests for
each patient, this whole cycle is repeated 9 times. It
means that the machine hence performs 1152 tests for
one complete batch. Every time a new parameter needs
to be estimated with reagent, it runs 6 standards and 2
controls before proceeding to the patient samples. By
testing the standards and comparing the results we are
calibrating the system every time each type of test
proceeds.
Sampling system: Sampling system consists of two
robotic arms and a stirrer arm. One of the robotic arms
handles the samples loaded in the sample tray and the
other arm handles the reagents loaded in the reagent
tray. The stirrer arm serves to stir the contents of the
reaction cell and also caters to washing and transporting
the reaction cell contents to micro flow cell for
1161
Res. J. Appl. Sci. Eng. Technol., 5(4): 1160-1168, 2013
Fig. 1: CAD drawing of analytical plate top view
Fig. 2: Block diagram of robotic arm
measurement purposes. Figure 2 depicts the block
diagram of arm function. Two different arms for
samples and reagents are provided to avoid cross
contamination between them. The sampling system
manages the loading, identification of each sample,
path manipulation of arm and transportation of blood
samples. The path of arms handling the reagents and
samples are manipulated in the flow diagram as shown
in Fig. 3 and 4. Both the arms are positioned in the
origin. Number of samples to be loaded is given as
count value of N, the number of reagent containers
loaded is given as count value of M.
When the process starts, the pneumatic pipettor
locks the pipette tip which is placed in the first reagent
container. It sucks 4 µL of reagent and transport to the
first reaction cell. The arm repeats the process with the
same reagent N times. For transfer of samples, the
pneumatic pipettor locks the pipette tip which is placed
in the first sample tube. It sucks 0.5 µL of sample and
transport to the first reaction cell. Then arm replaces the
pipette tip in the same sample tube and comes to the
origin position. Now the count value (N) = count
(N) - 1. The sample tray and reaction cell tray rotates
one step clock wise for the purpose of loading the next
sample in the next reaction cell. Next process of cycle
starts for the second sample and continues N number of
samples. All the time the reagent tray remains in the
fixed position as the reagent arm has accessed only one
particular reagent container. For one complete cycle,
the Reagent count (M) = Count (M) - 1. The sample
count is reloaded again as N for the next type of test.
The process then repeats for the second parameter with
all the samples i.e., N times.
The stirrer arm each time takes care of mixing,
incubating and transferring the incubated contents to
micro flow cell. All these arm and plate movements are
controlled by the micro controller and driven by the
servo motor. The value of counts loaded is done by
microcontroller programming. In 128 reaction cells 4
different types of tests for each patient can be
performed. The cleaning of whole reaction tray
subsequently needs to be done before the next 4 tests
are performed. For the 36 set of tests for each patient,
the whole process has to be repeated 12 times if the
numbers of samples loaded are 24. Based on the
number of samples, the testing cycle can be reduced
through the programming. The Robot Arm which is
1162
Res. J. Appl. Sci. Eng. Technol., 5(4): 1160-1168, 2013
Fig. 3: Transferring of reagent to reaction cells
Fig. 4: Transferring of samples to reaction cell
1163
Res. J. Appl. Sci. Eng. Technol., 5(4): 1160-1168, 2013
Fig. 5: Front view of sample/reagent arm
Fig. 6: Side view of sample/reagent arm
Table 1: Specification for analytical plate and arm function
Total number of tests
1152 tests/batch
Number of testing cycles
9 cycles/batch
Number of tests/cycle
128 tests
Sample tray capacity
0.8 mL tube, 32 samples
Micropipette for dispensing sample
0.1 mL resolution
Reagent tray capacity
25 mL tube, 36 reagents
Micropipette for dispensing reagent
1 and 0.2 ML resolution
Reaction cell tray capacity
1.2 mL, 128 cells
Temperature of reaction cell array
37°C, ±1°C resolution
handling the samples and reagents has a 4 Degrees of
Freedom (DOF) and the 3rd Arm is used to stir the
reaction cell content and move the contents to the
micro flow cell for the processing.
Each arm has separate pneumatic pipettor. The task
of handling each arm contains the movement of in and
out. It encompasses x-y-z or z-θ sample, reagent arm
movement and the conveyance of samples arm to the
reaction cells. Figure 5 and 6 shows CAD drawing of
Robotic arm in front and side view. Both the sampling
and Reagent arm looks the same. At the end of the arm,
the liquid handler, i.e., pneumatic pipettor is attached.
Table 1 shows the specification of analytical plate and
arm dimensions. This system can able to perform 36
tests per patient. So totally it runs up to 1152 tests per
batch. Controlling Unit consists of the micro controller
and all its associated circuitries. This unit controls the
entire system. For controlling the arm we are proposing
the scheduling algorithm.
temperature range of -5 to +70°C. It uses TTL/RS485
communication protocol for communicating with
Dynamixel servo motors.
The motors we have used are Dynamixel Servo
motors supplied by Robots (Cm 700 manual.com.
2012).They provide no load speed of 54 rpm and
holding torque of 24 kgf.cm at a 12 Volts DC supply.
They have an in-built STM32 series MCU that has the
ARM Cortex M3 architecture. These motors have
almost had a 0° to 360° full sweep. It also has a 12 bit
position sensor to depict the shaft position, providing a
feedback path for errors. It works on PID control
algorithm to enhance the precision. PID control actually
helps in considering the current error (P), integrate the
previous errors (I) and derive the value for future
errors (D).
Other than position feedback, it also gives
temperature and load values to prevent from
overheating and overloading. It has an analog resolution
of 4096 for 360° giving us a 0.08789° angle resolution.
It has a standby current of 100 mA and operating
current around 1.5 a, this reduces the power
consumption during no operation. Then the
communication between the STM32 (present in the
Dynamixel) and ATMega 2561 (present on CM700),
it’s a Half-Duplex Asynchronous serial communication.
All the motors are connected in a Daisy chain physical
link with the help of a daisy chain type connector. Then
each motor is assigned a specific ID to be identified in
the chain. This ID number can vary from 0×00 to 0×FD
giving us total 254 options. This ID can be saved in the
EEPROM of the motor MCU. Along with this data the
0 position, maximum temperature limit and other
necessary data about the motor are saved in the
EEPROM of STM32. Now whenever a specific angle is
to be obtained, ATMega 2561 board is programmed to
send the analog value along with the ID of motor where
shaft movement has to take place as a digital data
packet to STM32 MCU. The target motor controller
interprets the data packet and instructs DC motor to
move, with the help of the position sensor installed
inside the motor the high precision movement of DC
motor can be achieved and hence completing the
motion of the arm. Figure 7 shows the job scheduling
algorithm which describes the total functions of
Analytical Plate and Robotic Arm movement. Based on
this algorithm the Arm movement is manipulated
through the following equations:
N
∑ Re
P =1
N
N
= R1 + ∑ S N
(1)
N =1
Controlling unit: We are using a control module type
where,
controller with an ATMega 2561 MCU installed on it.
Re : Reaction cell
The operating voltage range is +7 to +35 V with an
R : Reagent container
overall maximum current ratings 10 A. There are
S : Sample tubes, for the first type of test of all the
temperature and voltage sensors added to the circuit for
patients
the data about the working atmosphere. It has a working
1164
Res. J. Appl. Sci. Eng. Technol., 5(4): 1160-1168, 2013
Fig. 7: Algorithm for sampler, reagent arm function and analytical plate rotation
Fig. 8: CM 700 development board with at Mega 2561 (cm 700 manual.com. 2012)
Likewise the following equations describes for the
complete test sequence:
2N
2N
= R2 + ∑ S N
3N
∑ Re
P =2 N +1
(2)
N
∑ Re
P = N +1
3N
N =1
=
(3)
N
R +∑S
2
N =1
N
From the above 3 Eq. (1), (2) and (3) we can
summarize the movement of plate and arm, in a
generalized equation:
( M +1) N
∑ Re
P = MN +1
( M +1) N
=
M +1
∑R
M =0
N
M +1
+ ∑SN
(4)
N =1
Eq. (4) denotes,
M : The total number of reagents used for one batch
process
N : The number of samples loaded
P : The number of reaction cells
ATMegha 2561 microcontroller unit has the RXD1 bit
2 and TXD1 bit 3 pins are connected with the daisy
chain link. MX 28 servo motor has ARM cortex M3
1165
Res. J. Appl. Sci. Eng. Technol., 5(4): 1160-1168, 2013
Fig. 9: Interfacing servomotors with microcontrollers in daisy chain link
Fig. 10: Time response of sample and reagent tray
processor, each processor receives data and sends to the
next motor because of daisy chain link is activated. The
motor has inbuilt sensory circuits that will detect the
temperature and sends data to the motor Fig. 8 shows
1166
Res. J. Appl. Sci. Eng. Technol., 5(4): 1160-1168, 2013
Fig. 11: Time response of reagent cells tray
the development board (CM 700) which has the control
module type controller and having a ATMegha 2561
processor.
RESULTS AND DISCUSSION
In this study the preliminary results of integration
between Arm and Analytical Plate has been presented.
The Robotic arm and Analytical Plate configurations
are user friendly and the reagent tray, which is
presented here, is easily removable. Three circular
plates have the different rotations and can be controlled
and programmed easily. Each time the number of
sample loaded can be varied up to 24. Based on the
sample size the operation can be adjusted automatically.
Figure 9 shows the Interfacing of servo motors with
microcontrollers. This board has the connection of
daisy chain link. With the daisy chain link we can able
to connect multiple servo motors one by one or in a ring
pattern. Each connecting motor acts as a master to the
next motor. It receives the data packets and sends the
information to the next. This system can able to run up
to 1152 tests/batch with 36 different kinds of tests. The
servo motor which is used here has an inbuilt ARM
cortex M3 processor. So it automatically corrects its
previous position and locates exactly where the cells are
presented. Figure 10 shows the result of time response
for sample and reagent tray rotations. This is the result
taken by running the sample tray and reagent tray with
simulated in 3D CAD Design Software Solid works.
Figure 11 shows the time response of reagent cells tray
this is also taken by simulating the reagent tray in 3D
CAD Design Software Solid works.
CONCLUSION
This study gives the overview and test results of
the Robotic Arm and Analytical Plate, which is
simulated, and the preliminary test results of the
Robotic Arm and Analytical Plate which is used in the
Clinical Chemistry Analyzer. These results are taken
from the 3D CAD Design Software Solid works. This
type of Arm and Plate can be used in the Complete
Automatic Analyzer that is used for small, medium
sized hospitals and the primary health centers. The
ultimate aim of this automated system is to bring out
1167
Res. J. Appl. Sci. Eng. Technol., 5(4): 1160-1168, 2013
the changes in human resource and precise use of
reagent by providing the automatic pipette tip. The
future work is carried out with the mathematical model
of the Robotic Arm, validating the synchronization
between the Robotic Arm Movement and Analytical
Plate and providing the menu driven software, which is
user friendly. The integrated system has to be
implemented in hospitals and validate system in real
time.
REFERENCES
Bhatia, P., J. Thiunarayanan and N. Dave, 1998. An
expert system-based design of SCARA robot.
Expert Syst. Appl., 15: 99-109.
Choi, B.J., K. Noh, J.W. Kim, S.M. Jin, J.C. Koo,
S.M. Ryew, J.K., W.H. Son, K.T. Ahn, W. Chung
and H.R. Choi, 2006. Intelligent biorobot platform
for integrated clinical test. International Joint
Conference, Oct., 18-21, Bexco, Busan, Korea, pp:
5828-5832, DOI: 10.1109/SICE.2006.315249.
Choi, B.J., S.M. Jin, S.H. Shin, J.C. Koo, S.M. Ryew,
M.C. Kim, J.H. Kim, W.H. Shon, K.T. Ahn,
W.K. Chung and H.R. Choi, 2008. Development of
flexible laboratory automation platform using
mobile agents in the clinical laboratory. IEEE
Conference on Automation Science and
Engineering, Suwon, pp: 918-923.
Choi, B.J., W.S. You, S.H. Shin, H. Moon, J.C. Koo,
W. Chung and H.R. Choi, 2011a. Robotic
laboratory automation platform based on mobile
agents for flexible clinical tests. IEEE Conference
on Automation Science and Engineering (CASE),
pp: 186-191.
Choi, B.J., W.S. You, S.H. Shin, H.M.J.C. Koo,
W. Chung and H.R. Choi, 2011b. Development of
robotic laboratory automation platform with
intelligent mobile agents for Clinical Chemistry.
IEEE Conference on Automation Science and
Engineering. Trieste, Italy, pp: 708-713.
Er, M.J., M.T. Lim and H.S. Lim, 2001. Real-time
hybrid adaptive fuzzy control of a SCARA robot.
Microprocess. Microsyst., 25: 369-378.
Nakamachi, E., 2010. Development of automatic
operated blood sampling system for portable type
self-monitoring blood glucose device. Conf. Proc.
IEEE Eng. Med. Biol. Soc., 2010: 335-338.
Saitoh, S. and T. Yoshimori, 2008. Fully automated
laboratory robotic system for automating sample
preparation and analysis to reduce cost and time in
drug development process. J. Assoc. Lab.
Automat.,
13:
265-274,
DOI:
10.1016/
j.jala.2008.07.001.
Taneja, S.R., R.C. Gupta, J. Kumar, K.K. Thariyan
and S. Verma, 2005. Design and development of
microcontroller-based clinical chemistry analyser
for measurement of various blood Biochemistry
parameters. J. Autom. Method Manag., 4: 223-229,
DOI: 10.1155/JAMMC.2005.223.
1168